University of Groningen
Talk to your gut
Espina, Marines du Teil; Gabarrini, Giorgio; Harmsen, Hermie J M; Westra, Johanna; van
Winkelhoff, Arie Jan; van Dijl, Jan Maarten
Published in:
FEMS Microbiology Reviews DOI:
10.1093/femsre/fuy035
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Final author's version (accepted by publisher, after peer review)
Publication date: 2019
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Espina, M. D. T., Gabarrini, G., Harmsen, H. J. M., Westra, J., van Winkelhoff, A. J., & van Dijl, J. M. (2019). Talk to your gut: The oral-gut microbiome axis and its immunomodulatory role in the etiology of rheumatoid arthritis. FEMS Microbiology Reviews, 43(1), 1-18. https://doi.org/10.1093/femsre/fuy035
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
Talk to your gut: the oral-gut microbiome axis and its
immunomodulatory role in the etiology of rheumatoid arthritis
Marines du Teil Espina1,*, Giorgio Gabarrini1,2,*, Hermie J.M. Harmsen1, Johanna Westra3,
Arie Jan van Winkelhoff1,2, and Jan Maarten van Dijl1,#
1University of Groningen, University Medical Center Groningen, Department of Medical
Microbiology, Hanzeplein 1, 9700 RB Groningen, the Netherlands
2University of Groningen, University Medical Center Groningen, Center for Dentistry and Oral
Hygiene, Antonius Deusinglaan 1, 9713 AV, Groningen, the Netherlands
3University of Groningen, University Medical Center Groningen, Department of
Rheumatology and Clinical Immunology, Hanzeplein 1, 9700 RB Groningen, the Netherlands;
*Both authors contributed equally to this work
One sentence summary: The microbiome as a driver in rheumatoid arthritis
Keywords: rheumatoid arthritis, microbiome, autoimmunity, Porphyromonas gingivalis, PPAD, neutrophils
Abstract
Microbial communities inhabiting the human body, collectively called the microbiome, are critical modulators of immunity. This notion is underpinned by associations between changes in the microbiome and particular autoimmune disorders. Specifically, in rheumatoid arthritis, one of the most frequently occurring autoimmune disorders worldwide, changes in the oral
#Corresponding author: J.M. van Dijl, Department of Medical Microbiology, University of Groningen, University Medical Center Groningen, Hanzeplein 1, P.O. box 30001, 9700 RB Groningen, the Netherlands, tel. +31-50-3615187, E-mail: j.m.van.dijl01@umcg.nl
and gut microbiomes have been implicated in the loss of tolerance against self-antigens and in increased inflammatory events promoting the damage of joints. In the present review, we highlight recently gained insights in the roles of microbes in the etiology of rheumatoid arthritis. In addition, we address important immunomodulatory processes, including biofilm formation and neutrophil function, which have been implicated in host-microbe interactions relevant for rheumatoid arthritis. Lastly, we present recent advances in the development and evaluation of emerging microbiome-based therapeutic approaches. Altogether, we conclude that the key to uncovering the etiopathogenesis of rheumatoid arthritis will lie in the immunomodulatory functions of the oral and gut microbiomes.
Introduction
The many trillions of microbes we harbor in our bodies are not pure spectators. Indeed, they play a fundamental role in shaping our immune system and metabolism as has become increasingly evident in recent years (Cho & Blaser, 2012; Belkaid & Hand, 2014; Muszer et al., 2015; Lynch & Pedersen, 2016; Levy et al., 2017). These microbes, which altogether constitute our microbiome, are located in the gastrointestinal tract, the nose, the oral cavity, the skin, the vagina, and, to a lesser extent, the lungs (Cho & Blaser, 2012; Muszer et al., 2015). Interestingly, compositional changes of the microbiome, altogether categorized as dysbiosis (Cho & Blaser, 2012), have been associated with a broad range of diseases including metabolic and autoimmune disorders (Cho & Blaser, 2012; Muszer et al., 2015;
Levy et al., 2017). Since then, efforts have been made to define a “healthy microbiome”, but
only as of late, with the use of sophisticated sequencing technologies and computational methods for data analysis, bountiful progress has been made in this field (Manor et al., 2014;
Wang et al., 2015). One important example of this progress is the Human Microbiome
Project (Turnbaugh et al., 2007; Human Microbiome Project Consortium, 2012; Lloyd-Price
et al., 2017), implemented by the US National Institutes of Health. The large-scale
high-throughput analyses performed in this project yielded over 350 papers providing important clues on how the microbiome and its expressed genes play a role in health and disease
(Muszer et al., 2015). Dysbiotic conditions have therefore been the subject of critical studies, especially to uncover factors leading to this unbalance of the complex status quo in which microbial communities interact within and with the human body. Factors altering microbial homeostasis include the use of antibiotics and other drugs, changes in diet patterns, elimination of constitutive nematodes, the introduction of a new microbial actor, and ageing (Cho & Blaser, 2012; Belkaid & Hand, 2014; Lynch & Pedersen, 2016; Levy et al., 2017; Buford, 2017; Imhann et al., 2017; Maier et al., 2018).
Intriguingly, despite the associations between microbiome and autoimmunity, the tissue targeted by autoimmune disorders is often not the same tissue where the microbiome is thought to exert its pathogenic role (Brusca et al., 2014; Chen et al., 2017). This is clearly exemplified by rheumatoid arthritis (RA), one of the most prevalent autoimmune diseases, affecting approximately 1% of the human population (Potempa et al., 2017). RA thus contributes significantly to the global morbidity and mortality and, according to the allegations of its increasingly higher incidence among the elderly population (Tutuncu & Kavanaugh, 2007; Kato et al., 2017), it is a major threat to healthy ageing (Janssen et al., 2013; van Onna & Boonen, 2016). RA is characterized by a persistent synovial inflammation, which ultimately results in articular cartilage and bone damage (McInnes & Schett, 2011). Recent models have implicated the involvement of loss of tolerance toward citrullinated proteins in RA development (de Smit et al., 2011; Quirke et al., 2014; Darrah & Andrade, 2017). Citrullination is a post-translational protein modification involving the transformation of a positively charged arginine residue into a neutral citrulline residue (de Smit et al., 2011). This reaction is catalyzed by peptidylarginine deiminase (PAD) enzymes, which are extremely well conserved among mammals (Gabarrini et al., 2015). Of note, human PAD enzymes regulate, in a variety of cells and tissues, important processes such as apoptosis, inflammatory immune responses, and the formation of rigid structures like skin or myelin sheaths (Mangat et al., 2010; Wegner et al., 2010; Witalison et al., 2015). Consistent with RA etiological models, in the majority of predisposed subjects, the presence of citrullinated
proteins gives rise to specific autoantibodies called anti-citrullinated protein antibodies (ACPAs) (Pischon et al., 2008; Detert et al., 2010; Quirke et al., 2014). Remarkably, ACPAs have a specificity of 95% and are 68% sensitive for RA (Avouac et al., 2006; Nishimura et al., 2007). These auto-antibodies can be detected years before the appearance of clinical symptoms (Arkema et al., 2013). Moreover, their serum levels strongly correlate with disease severity, hinting at a possible role in the progression of the disease (Koziel et al., 2014).
The etiology of RA is still not fully understood but, among its potential causes, certain genetic factors were shown to strongly correlate with the disease. Particularly, the major histocompatibility complex (HLA)-DRB1 locus is one of the most well-established genetic risk factors associated with RA and ACPAs (McInnes & Schett, 2011). Specifically, alleles coding for a five amino acid sequence called shared epitope, which is present in the HLA-DRB1 region, are carried by 80% of ACPA+ RA patients (Huizinga et al., 2005) and correlate with
disease activity and mortality (Farragher et al., 2008; Nagafuchi et al., 2016) (Fig. 1). The shared epitope appears to favor the binding of citrulline-containing peptides during HLA presentation when compared to their non-citrullinated counterparts, although this hypothesis seems to be applicable only to certain shared epitope alleles such as HLA-DRB1*04:01, *04:04 and *04:05 (Kampstra et al., 2016; Scally et al., 2017). Nevertheless, it appears that the genetic component is only one of the many RA-contributing factors. Specifically, environmental ones have always attracted great attention for multiple reasons. In particular, it is noteworthy that the genetic component is not sufficient to explain the recent increase in RA prevalence among the population (Myasoedova et al., 2010). An additional, more intuitive, reason is that not every individual carrying the alleles implicated in RA susceptibility develops RA (Holoshitz, 2013). Important clues for the identification of environmental triggers of RA were provided in the beginning of the 20th century, when treatment of
periodontal infections were proven to ameliorate symptoms of patients with rheumatoid arthritis (Sturridge, 1918). Since then, it has become increasingly more evident that oral
health and especially the oral microbiome significantly influence the progression of RA (Diamanti et al., 2016; Potempa et al., 2017). Studies consistent with this line of thought revealed another, less apparent, actor playing a role in the pathogenesis of RA: the gut microbiome (Scher et al., 2016a) (Fig. 1). Indeed, Zhang et al. recently analyzed the microbiome composition of fecal, dental, and salivary samples of RA patients, showing that both the oral and gut microbiomes were dysbiotic compared to the ones of healthy individuals (Zhang et al., 2015). Strikingly, the dysbiotic characteristics were shown to be partially resolved after RA treatment, which implied an interplay between RA and the oral-gut axis (Zhang et al., 2015). Understanding the role of the microbiome in RA is therefore essential to fully understand the etiopathological landscape of RA. Additionally, this insight might also be useful in understanding similar, related, autoimmune diseases such as systemic lupus erythematosus (SLE) (de Oliveira et al., 2017). In this review, we discuss the most relevant findings on how the interplay of both the oral and gut microbiomes with the host mediate RA onset, focusing on recently proposed factors such as biofilms and neutrophil function. Lastly, we will address how this information could eventually lead to the identification of potentially druggable targets for a microbiome-based therapeutic management of RA and other autoimmune diseases.
Oral microbiome, periodontitis and RA
Oral health has been clinically associated with autoimmune diseases in a number of epidemiological studies (Mercado et al., 2000; Pischon et al., 2008; de Pablo et al., 2008; de
Smit et al., 2012; Correa et al., 2017; Rusthen et al., 2017) (Tables 1 and 2). An important
example of this is the correlation between RA and periodontitis, which is a chronic inflammatory disorder affecting the periodontium, the tissue supporting the teeth (Mercado et al., 2000). Periodontitis is a major cause of tooth-loss and one of the most widespread diseases in the world, with an incidence of roughly 11% in the human population (Potempa
et al., 2017), although the disease affects between 10 to 57% of different populations
worldwide, depending on severity, socio-economic status, and oral hygiene (Rylev & Kilian,
2008). As mentioned, a recent cause of concern for this disease is its long-known correlation with RA (Pischon et al., 2008; de Pablo et al., 2008). It has been reported, in fact, that periodontitis patients have twice the chance of contracting rheumatoid arthritis and RA patients are twice as likely to become edentulous (Kasser et al., 1997; Mercado et al., 2000; Mercado et al., 2001; Pischon et al., 2008; Dissick et al., 2010). Additionally, treatment of periodontitis has been shown to ameliorate symptoms of rheumatoid arthritis and vice versa
(Ribeiro et al., 2005; Al-Katma et al., 2007; Pers et al., 2008; Ortiz et al., 2009), and the citrullinome of periodontopathic conditions mirrors the one of the arthritic inflamed joint (Konig et al., 2016). However, the molecular mechanism behind this association has not yet been elucidated. Nevertheless, strong evidence suggests that RA autoimmunity is triggered or enhanced by specific oral bacteria that are causatives of periodontal disease (Wegner et al., 2010; Hitchon & El-Gabalawy, 2011; de Smit et al., 2012; Maresz et al., 2013; Konig et al., 2016). The Gram-negative bacterium Porphyromonas gingivalis is the main suspect in the association between periodontitis and RA (Janssen et al., 2013). This was firstly due to the fact that antibody responses against P. gingivalis and specific P. gingivalis virulence factors appeared to correlate with RA severity and ACPA levels (Tolo & Jorkjend, 1990; Hitchon et al., 2010; Kharlamova et al., 2016), even more strongly than with smoking, a well-known RA risk factor (Kharlamova et al., 2016). Secondly, in more recent times, a peculiar
P. gingivalis enzyme has been hailed as the lynchpin of the link between periodontitis and
RA (Rosenstein et al., 2004). This protein is the PAD enzyme of P. gingivalis (PPAD), the only thus far reported PAD enzyme produced by a human pathogen (Gabarrini et al., 2015; Gabarrini et al., 2018a; Gabarrini et al., 2018b). Antibodies against PPAD, in fact, have been shown to correlate with RA in several studies (Quirke et al., 2014; Shimada et al., 2016). Albeit contradicting observations have been made (Konig et al., 2015; Zhang et al., 2015), PPAD involvement in RA development was implied by experimental studies in RA murine models (Maresz et al., 2013; Gully et al., 2014). In these studies, either genetically engineered PPAD-deficient P. gingivalis mutants or the wild-type strains were used to infect mice in which arthritis was experimentally induced. A higher autoantibody production as well
as higher joint damage were observed in mice infected with the wild-type strain compared to the ones infected with PPAD-deficient mutants, suggesting a role for PPAD in the exacerbation of RA. This bacterial enzyme is evolutionary unrelated to mammalian PADs, but it nonetheless shares with this group of eukaryotic enzymes the catalytic function (Koziel
et al., 2014). Of note, PPAD is purported to play a role in RA etiology with two potential
mechanisms. The first one requires the proteolytic activity of a specific class of highly efficient proteases secreted by P. gingivalis, named arginine-gingipains, which were shown to be necessary for α-enolase citrullination (Wegner et al., 2010). In vitro experiments showed that cleavage of host proteins by gingipains, in fact, exposes carboxyl-terminal arginine residues, which are the preferential targets of PPAD (McGraw et al., 1999; Wegner
et al., 2010). This unique mode of citrullination of cleaved peptides may be the basis of the
generation of so-called neo-epitopes at sites where PPAD activity has been suggested, such as the sites of infection or even distant periodontal tissues (Laugisch et al., 2016). Neo-epitopes are Neo-epitopes to which immune tolerance has not yet been developed, consequently triggering an autoimmune response (Wegner et al., 2010) (Fig. 2). The second mechanism involves molecular mimicry (Fig. 2). It has been shown, in fact, that autoantibodies directed against the immunodominant epitope of human citrullinated α-enolase cross-react with P.
gingivalis citrullinated α-enolase (Lundberg et al., 2008). These observations were further
confirmed by Li et al., who additionally identified six P. gingivalis citrullinated peptides recognized by RA-derived ACPAs (Li et al., 2016). Besides the hypotheses proposing a causative relationship between PPAD production and RA autoimmunity, however, other oral microbiome-driven mechanisms mediating loss of tolerance against citrullinated proteins have been proposed. The first is enhanced human PAD-mediated citrullination (Maresz et al., 2013). Inflammatory processes that can be triggered by microbial events, in fact, have been known to involve PAD-mediated citrullination. In the case of chronic inflammations, such as periodontitis, continuous PAD activation might lead to an enhanced citrullination burden and, potentially, autoimmunity (Nesse et al., 2012; Valesini et al., 2015) (Fig. 2). Dysbiosis is therefore considered to be a critical driver for the perpetuation of inflammatory
statuses and break in tolerance against citrullinated proteins (Darveau et al., 2012; Holers, 2013). Interestingly, P. gingivalis, albeit underrepresented in the periodontal oral microbiome, appears to be capable of causing inflammatory responses by orchestrating oral dysbiosis (Hajishengallis et al., 2012; Maekawa et al., 2014). This peculiar feat, which placed
P. gingivalis in the limelight as a “keystone pathogen”, creates a suitable environment for
dysbiotic bacteria to persist, aggravating the loop between oral dysbiosis and inflammation (Hajishengallis et al., 2012) (Fig. 2).
Besides a direct or indirect modulation of citrullination, the oral microbiome influences other processes, mainly involving the T cell-mediated adaptive immunity, that have been correlated with chronic inflammation and bone damage in the RA joints (Noack & Miossec, 2014; de Vries et al., 2017). Specifically, T helper 17 (Th17) cells, a subset of CD4+ T cells
normally produced against bacterial or fungal infections, have been associated with joint damage via mechanisms such as overproduction of the proinflammatory cytokines IL-17A, IL-17F, and IL-22, cross-reactivity with joint-derived antigens, or migration to the joints, where increased osteoclast activation mediates bone resorption (Sato et al., 2006; Hot et al., 2011; Zhao et al., 2013; Rogier et al., 2015). These pathological Th17 cells can be produced in the oral cavity in response to certain periodontal pathogens (Moutsopoulos et al., 2012; de Aquino et al., 2014; Tsukasaki et al., 2018). Accordingly, Th17 cells and Th17-related cytokines are often observed in ex vivo gingival tissue samples of periodontitis patients (Moutsopoulos et al., 2012). Additionally, a recent study using a periodontitis mouse model was characterized by accumulation of, among CD4+ T cell subsets, only Th17 cells. This accumulation was reverted after administration of antibiotics, corroborating the hypothesized role of the oral microbiome in the production of Th17 cells and their ensuing responses (Tsukasaki et al., 2018). Accordingly, P. gingivalis was shown to specifically induce the production of Th17-related cytokines in vitro, a mechanism that involved gingipain degradation of specific cytokine mediators that favored Th17 responses (Moutsopoulos et al., 2012). Moreover, it was later confirmed in collagen-induced arthritis (CIA) mice, that
induction of periodontitis by P. gingivalis and another Gram-negative bacterium, Prevotella
nigrescens, resulted in increased presence of Th17 cells in lymph nodes draining arthritic
joints, and in aggravation of arthritic symptoms (de Aquino et al., 2014). The mechanisms by which Th17 responses are enhanced by these two oral pathogens involved IL-1 activity and the activation of antigen-presenting cells via Toll Like Receptor 2 (de Aquino et al., 2014). Additional, less explored, mechanisms underlying the interplay of P. gingivalis and RA etiology are further detailed in particular dedicated sections of this review.
In recent years, studies investigating RA pathogenesis have implicated other periodontal pathogens aside from P. gingivalis in this disease. Schwenzer et al. demonstrated that the serological response against Prevotella intermedia in RA patients was associated with a novel ACPA directed against cCK13‐1, a newly discovered citrullinated peptide of cytokeratin 13, found in the periodontium (Schwenzer et al., 2017). Interestingly, unlike other ACPAs, this autoantibody did not correlate with a serological response against P. gingivalis, suggesting that ACPAs with different specificities might arise from responses to different oral periodontal pathogens (Schwenzer et al., 2017).
Another study, has recently implicated the Gram-negative bacterium Aggregatibacter
actinomycetemcomitans in the etiology of RA through the enhancement of citrullination
(Konig et al., 2016). The mechanism behind this purported association appears to depend on the pore-forming leukotoxin of A. actinomycetemcomitans, LtxA. Upon a lytic stimulus from this toxin, destruction of the neutrophil membrane occurs, thus releasing human PADs and leading to hypercitrullination (Hajishengallis et al., 2012) (Fig. 2). A correlation between LtxA and RA was further demonstrated, as anti-LtxA antibodies were associated with ACPA serum titers in RA patients. The biomolecular rationale behind this mechanism is further explained in the “Neutrophils and RA pathogenesis” section below.
Aside from the aforementioned studies, which have investigated the involvement of specific oral species in RA etiopathogenesis, efforts have been made to analyze the oral microbiome
composition in RA patients. Scher et al. 2012 analyzed the microbial composition of rheumatoid arthritis and control patients with and without periodontitis. New-onset RA patients (NORA), chronic RA (CRA) patients, and healthy control volunteers were included in this study, in order to pinpoint specific bacteria that are associated with different stages of RA progression. Among all groups analyzed, NORA patients exhibited high incidence of advanced periodontal disease. Intriguingly, the microbial richness and composition did not show a significant variation among all groups with a similar periodontitis status (Scher et al., 2012). However, two taxa of Gram-negative bacteria were exclusively found in NORA patients irrespective of periodontal disease, namely Prevotella spp. and Leptotrichia spp. (Scher et al., 2012). Moreover, ACPA levels were positively associated with the presence and abundance of yet another Gram-negative bacterium, Anaeroglobus geminatus, indicating a possible role of this bacterium in RA initiation. An unexpected finding was that presence and abundance of P. gingivalis was not positively associated with RA or with ACPA serum titers, but only with periodontitis severity (Scher et al., 2012).
Zhang et al. 2015, on the other hand, analyzed fecal, dental and salivary samples of RA patients observing a dysbiotic gut and oral microbiota compared to healthy individuals. Particular attention was given to Gram-negative bacterial Haemophilus species, which were underrepresented in the oral and gut compartments of RA patients and which negatively correlated with autoantibodies related to RA (Zhang et al., 2015). In contrast, the Gram-positive Lactobacillus salivarus was overrepresented in all body sites tested of RA patients and positively correlated with disease activity (Zhang et al., 2015). Lopez-Oliva et al. also analyzed the oral microbiome composition in periodontally healthy individuals with or without RA. Similarly to Zhang et al., the study showed that the microbiome of RA patients is enriched for certain Gram-negative species with proinflammatory capacity including
Prevotella spp. and, similarly to Scher et al., Leptotrichia spp., suggesting a possible role for
these two bacteria in the initiation of RA (Scher et al., 2012; Lopez-Oliva et al., 2018). Additionally, the Gram-positive Cryptobacterium curtum was identified as the predominant
species in the microbiome of RA patients (Lopez-Oliva et al., 2018). This is of interest particularly due to C. curtum‟s capability of citrullinating free arginine through the arginine
deiminase pathway, albeit ACPAs target citrullinated proteins and not free citrulline.
Another recent study (Mikuls et al., 2018) investigated the subgingival microbiome of RA patients using as control the microbiome of osteoarthritis (OA) patients, in order to pinpoint specific correlations with the autoimmune side of rheumatoid arthritis. Interestingly, after taking the periodontal status into account, no robust microbial fingerprint was found for RA when compared to OA (Mikuls et al., 2018). Remarkably, in contrast with previous studies, no correlation was observed between serum ACPA levels and abundance of bacteria that have been associated with RA, such as P. gingivalis, A. germinatus, Haemophilus, or
Aggregatibacter (Mikuls et al., 2018). Additionally, an under-representation of
Peptostreptococcus, Porphyromonas, Prevotella and Treponema species was observed in
RA patients with periodontitis compared to OA patients with periodontitis. Of note, early RA patients also presented an under-representation of Prevotella and Porphyromonas species in the microbiome of their lung, which has recently emerged as another important extra-articular site where RA autoimmunity may develop (Scher et al., 2016b). It must be noted that all the aforementioned metagenomic studies are correlational and therefore do not necessarily imply involvement of specific bacteria in the causation of a disease. Moreover, abundance of a microbe does not always correlate with the serological host response (de
Smit et al., 2012), a factor known to have implications in ACPA formation and RA (de Smit et
al., 2012; Li et al., 2016; Schwenzer et al., 2017). The results of these studies, however, albeit not giving insights into the mechanisms behind the pathogenesis of RA, might lead to advancements in diagnosis and prognosis of this disease.
Oral biofilms and their role in inflammatory responses
One of the factors leading to chronic inflammation in periodontitis is the dental biofilm, which consists of highly complex, organized, and dynamic microbial communities that acquire in this way resistance to environmental stresses, including antibiotics (Zijnge et al., 2010; Kostakioti et al., 2013; Sintim & Gursoy, 2016) (Fig. 3). For this reason, biofilms have been studied thoroughly by the medical community. However, direct correlations between biofilms and autoimmune disorders have not yet been examined extensively. Nevertheless, in recent years, studies showed the immuno-modulating properties of specific biofilm mediators responsible for biofilm formation. Some of such mediators are called autoinducers, the most conserved signaling molecules that allow “communication” among bacteria in an interconnecting process known as quorum sensing (Kostakioti et al., 2013). Specifically, the quorum sensing molecule autoinducer 2 (AI-2), which can be secreted and sensed by both Gram-positive and Gram-negative bacteria, has been shown to mediate the virulence and biofilm formation of periodontal pathogens (Kolenbrander et al., 2010). Moreover, the involvement of AI-2 in inflammation processes was recently demonstrated by Zargar et al.,
who analyzed the transcriptome of human intestinal epithelial cells (IECs) when exposed to proteins secreted by two strains of non-pathogenic Escherichia coli that differ mainly in their production of AI-2. The differential inflammatory response of the IECs prompted the authors to study the specific role of AI-2 by stimulating these cells with synthetic AI-2 (Zargar et al., 2015). Their results suggest that IECs are able to alter the transcription of immune mediators, such as the neutrophil-recruiting interleukin 8 (IL-8), when faced with quorum sensing molecules. Moreover, in the case of RA and its alleged relationship with periodontitis, AI-2 molecules have been demonstrated to mediate oral biofilm formation (Sintim & Gursoy, 2016). Furthermore, AI-2 molecules were found to be expressed by periodontal pathogens, such as P. gingivalis (Chung et al., 2001; Fong et al., 2001), belonging to one of the nine taxa composing the main core of dental biofilms, termed hedgehog structure (Kriebel et al., 2018). Of note, this bacterium has been found capable of
inducing secretion of IL-8 in oral epithelial cells (Yee et al., 2014). This observation is apparently supported by the finding that addition to oral bacteria of the “red complex”, a group of bacteria comprising P. gingivalis, Tannerella forsythia, and Treponema denticola, can increase the IL-8 production in oral epithelial cells (Belibasakis et al., 2013; Zargar et al., 2015). Taken together, these results suggest that quorum sensing signaling molecules released by P. gingivalis might lead to an inflammatory response in the oral cavity, as schematically represented in Figure 2. As biofilms have been proven to represent inflammation-causing etiological factors of periodontitis, studies delving deeper into this topic might also help widening our understanding of the periodontal field and potentially inflammation-driven autoimmunity.
Neutrophils and RA pathogenesis
Another suggested oral microbiome-mediated mechanism potentially contributing to autoimmunity involves neutrophils. Neutrophils act as a first line of defense in periodontal diseases and are important regulators of both innate and adaptive immunity. Aberrant neutrophil functions have recently emerged as actors in the initiation and pathogenesis of autoimmune diseases, such as SLE and RA (Nemeth & Mocsai, 2012; Wright et al., 2014). In SLE, neutrophils exhibit impaired phagocytosis, have a tendency to form aggregates, display an elevated apoptotic behavior and an increased activation state mediated by nucleosomes (Abramson et al., 1983; Wu et al., 2013). Similarly, in RA, an increased recruitment and activation of neutrophils in synovial fluid can be observed in the early stages of the disease (Kaplan, 2013). A similar neutrophil phenotype is present in periodontitis where it has been suggested that neutrophils are key players in the initiation and perpetuation of the inflammation of gingival tissue (Lakschevitz et al., 2013). Interestingly, one important neutrophilic mechanism, which was recently strongly correlated with autoimmunity in SLE and RA, is the production of neutrophil extracellular traps (NETs), a process also known as NETosis. The NETosis mechanism involves the release, after the lysis of a neutrophil, of decondensed chromatin with, bound to it, a variety of proteins such
as histones and antimicrobial peptides, forming traps for targeted microorganisms (Wright et al., 2014). One essential step in NETosis is the citrullination of nuclear proteins such as histones (specifically H2A, H3, and H4) by the PAD4 enzyme, which is abundantly expressed in neutrophils (Nakashima et al., 2002) (Fig. 4). Notably, in healthy individuals, NETs are cleared after they have performed their extracellular killing function. The clearance of NETs is known to be performed by extracellular DNAse I and by macrophages, which can engulf and digest NETs (Hakkim et al., 2010; Farrera & Fadeel, 2013). This function might be fundamental, considering that release into the extracellular environment of NET components, such as citrullinated proteins and DNA, might create a potential new source of autoantigens in genetically susceptible hosts, thereby potentially contributing to autoimmune diseases (Khandpur et al., 2013) (Fig. 2). In fact, autoantibodies against citrullinated histone H4, H2A and H2B are commonly found in RA patients (Pratesi et al., 2014; Corsiero et al., 2016). Moreover, incomplete clearance of NETs, coupled with chronic inflammation, has already been correlated with initiation and/or development of autoimmune responses toward DNA and citrullinated proteins (Hakkim et al., 2010; Radic, 2014). Additionally, aside from an incomplete clearance of NETs, another reported factor leading to the production of autoantibodies against intracellular antigens is enhanced NETosis (Khandpur et al., 2013). Examples of the effects of these two aberrant NETosis events may be encountered in SLE and RA. In SLE, incomplete clearance of NETs has been observed, potentially due to the presence of DNAse I inhibitors or anti-NET antibodies that prevent DNAse I to break down NETs (Hakkim et al., 2010). In contrast, enhanced NETosis was observed in the synovial fluid of RA patients (Khandpur et al., 2013). Both events are likely to result in a constant stimulation of the immune system leading to autoantibody production (Wright et al., 2014).
While these findings suggest that neutrophils are essential players in the pathogenesis of periodontitis and RA, several studies investigated the relation of these cells to the oral microbiome (Uriarte et al., 2016). This was especially due to the fact that specific members of this microbiota, such as A.actinomycetemcomitans and P. gingivalis, are able to mediate
immune mechanisms relevant for RA (Konig et al., 2016). One of such mechanisms is the aforementioned hypercitrullination (Romero et al., 2013). This relates to the fact that human PADs require calcium cations to perform their citrullinating function (Darrah & Andrade, 2017). Consequently, when human PADs are released into the calcium-rich extracellular space after a NETosis event, they will tend to exert an increased and aberrant citrullinating function (Romero et al., 2013; Darrah & Andrade, 2017). A particular mechanism for the release of PADs in the Ca2+-rich extracellular milieu is observed in a study of A.
actinomycetemcomitans, in which the bacterial toxin LtxA was shown to cause membrane
lysis, potentially leading to the hypercitrullination scenario observed during NETosis (Konig
et al., 2016) (Fig. 2). P. gingivalis lipopolysaccharide, on the other hand, was shown to inhibit
apoptosis of neutrophils (Murray & Wilton, 2003) and increase epithelial secretion of IL-8
(Yee et al., 2014), an act that stimulates neutrophil migration towards the periodontal tissue
and into the gingival crevice (Fig. 2). These events are of particular interest, since lifespan prolongation and increased migratory behavior of neutrophils can lead to an augmented and persistent immune response, which is the ideal condition for the onset of an autoimmune reaction. An additional piece of evidence for the relation between oral microbiome, neutrophils, and RA is given by several observations showing that neutrophils, in periodontal pockets, employ mainly NETosis as a defense mechanism against periodontal pathogens (Delbosc et al., 2011; Kaplan, 2013; White et al., 2016). Clearly, since the molecular background of NET formation is still largely unknown, many possible hypotheses on the roles of this process remain to be evaluated, especially in vivo where the interplay between multi-species biofilms and the host can lead to the biological outcomes observed in autoinflammatory diseases.
Microbial translocation to the joints
The oral cavity is not the only location where oral bacteria, and especially P. gingivalis, have been thought to exert a pathogenic activity. Translocation of oral bacteria to other body compartments has been evidenced as well (Aagaard et al., 2014; Blanc et al., 2015; Gao et
al., 2016; Mougeot et al., 2017). While the mechanisms used by these bacteria, or their components, to reach distant locations in the body have not been completely elucidated, several hypotheses have been postulated. For instance, a direct entry of oral bacteria into the bloodstream has been proven during common dental practices, such as teeth brushing, flushing, and mastication (Li et al., 2000; Parahitiyawa et al., 2009). This entry mechanism seems to be enhanced under inflammatory conditions, such as periodontitis, due to higher proliferation and dilation of the periodontal vasculature. A recent study showed that oral bacteria could be found in the liver and spleen of mice with experimentally induced periodontitis. Notably, this bacterial translocation stopped after tooth removal and healing of gingival tissues, suggesting that periodontal bacteria can disseminate during breakdown of the oral barrier (Tsukasaki et al., 2018). Another proposed mechanism of bacterial translocation is the use of host cells as a „Trojan horse‟ (Hajishengallis, 2015). P. gingivalis, in fact, is known to survive intracellularly within several cell types, such as macrophages and dendritic cells, both of which can subsequently enter the blood stream and have the potential to disseminate bacteria throughout the body (Singh et al., 2011; Carrion et al., 2012). Microbial translocation is of interest in the context of RA, due to the fact that immune activation mechanisms and local inflammation could occur in response to the presence of oral bacteria or their components in the synovial joints (Farquharson et al., 2012; Chukkapalli et al., 2016) (Fig. 1). Corroborating this hypothesis, several studies have demonstrated the presence of P. gingivalis, P. intermedia and F. nucleatum DNA in the synovial fluid of RA patients with periodontitis (Martinez-Martinez et al., 2009; Temoin et al., 2012; Totaro et al., 2013). P. gingivalis DNA has also been found in the joints of a murine collagen-induced arthritis model infected with “red complex” bacteria. Perhaps more remarkably, the presence of DNA of oral bacteria in the synovial joints was found to associate with arthritis exacerbation (Chukkapalli et al., 2016). Although translocation of viable oral bacterial cells to the joint compartment of RA patients appears to be plausible, as shown for atherosclerotic plaques (Kozarov et al., 2005), transport of bacterial components
is a more supported hypothesis to explain the presence of genetic material of P. gingivalis
and other oral bacteria in the joints of RA patients (Martinez-Martinez et al., 2009).
The influence of the oral microbiome on the gut in the context of RA
Albeit infrequently, oral bacteria have been implicated in non-oral infections, among them intra-abdominal and intra-cranial sites, the appendix, and the lungs (van Winkelhoff & Slots, 1999). In particular, mobilization of bacteria from the oral compartment to distant physiological sites may involve the gastrointestinal tract, as humans continuously swallow oral bacteria (Arimatsu et al., 2014; Nakajima et al., 2015; Flynn et al., 2016). This view is supported by the incidental detection of oral bacterial DNA in human feces samples (Harmsen HMJ, unpublished observations). In recent years, the impact of the oral microbiome on the gut microbial composition has been investigated in several disease scenarios. Intriguingly, oral bacteria, including P. gingivalis, A. actinomycetemcomitans, and
Fusobacterium spp. have been implicated in several gastrointestinal diseases including
pancreatic and colorectal cancer (CRC) (Klimesova et al., 2018). Moreover, Atarashi et al., showed in germ-free mice that oral bacteria are capable of colonizing the gut, causing chronic inflammatory reactions in predisposed hosts (Atarashi et al., 2017). In the case of P.
gingivalis, however, Geva-Zatorsky et al. recently suggested that this bacterium is not
capable of colonizing the gut of germ-free mice since it could not be cultured from the feces of these mice (Geva-Zatorsky et al., 2017). Nonetheless, there is a possibility that this bacterium could remain viable during its passage through the acidic environment of the stomach due to its strong resistance to acid (Sato et al., 2017) and therefore, under proper conditions, potentially establishes a foothold in the human intestine. Interestingly, oral administration of P. gingivalis had significant repercussions on the bacterial composition of the gut, specifically decreasing the proportion of Bacteroidetes and increasing the proportion of Firmicutes (Sato et al., 2017). This, together with the fact that P. gingivalis was found to lower the complexity of gut bacterial communities, suggests a role for this bacterium in modulating the gut microbiome (Kramer et al., 2017). Similarly, oral administration of A.
actinomycetemcomitans was shown to modulate the gut microbiome of mice, a process that was correlated with metabolic and immunological changes involved in non-alcoholic fatty liver disease (Komazaki et al., 2017). The implications of the above-mentioned observations are highly relevant, considering that shifts in the gut microbial composition have been shown to have a significant impact on autoimmune disorders such as RA (Wu & Wu, 2012). In a murine collagen-induced arthritis model, in fact, P. gingivalis administration significantly changed the gut microbiome, while it simultaneously increased Th17 responses and aggravated arthritis (Sato et al., 2017). An association between the oral and gut microbiomes was, instead, recently described for RA patients in one of the aforementioned studies (Zhang et al., 2015). These individuals presented simultaneously dysbiotic oral and gut microbiomes, with decreased oral levels of Haemophilus spp. and increased levels of
Lactobacillus salivarius in the gut, both of which were partially resolved after RA treatment
(Zhang et al., 2015). Taken together, these findings suggest that bacteria belonging to the oral microbiome are capable of disrupting the eubiotic state of the gut microbiome, an act that can lead to chronic inflammation and trigger or enhance RA (Fig. 1).
Gut microbiome and RA
Mucosal sites are constantly exposed to microbial challenge and are considered of great importance in the initiation and modulation of microbiome-induced inflammatory responses. Among these sites, the gut is the one that has attracted most attention in the modulation of the host metabolism and immunity, due to its massive colonization by microorganisms (Belkaid & Hand, 2014). Dysbiosis of the gut microbiome has been shown to be related to RA pathogenesis by several studies (Scher et al., 2016a). Recently, Dorozynska et al.
demonstrated that a partial depletion of the natural gut microbiota due to antibiotics aggravated arthritis symptoms in an RA murine model (Dorozynska et al., 2014). As for the previously observed correlation between oral microbiome and RA, the gut microbiome is also linked to RA through T cell mediated immunity. In healthy individuals, a CD4+ T cell
subtype known as regulatory T cells (Tregs) is in balance with Th17 cells, and has an
inflammatory role that prevents the onset of autoimmune responses (Noack & Miossec, 2014) (Fig. 5A). Interestingly, a recent study detected decreased levels of Tregs and elevated levels of Th17 cells in the peripheral blood of RA patients, hinting at a Th17/Treg imbalance in these patients (Wang et al., 2012). Notably, this balance can be altered by the gut microbiome, considering that production/reduction of either Th17 cells or Tregs can be orchestrated by the gut microbiota via Toll-like receptor 2 (TLR2) (Nyirenda et al., 2011). Indeed, Tregs expressing the transcription factor Foxp3 (also called Foxp3+ Tregs) are
known to promote homeostasis and were found to have an increase in population size dictated by TLR2 sensing of the polysaccharide A (PSA) of Bacteroides fragilis, a human symbiont (Round & Mazmanian, 2010) (Fig. 5B). Moreover, Tregs expressing the hormone receptor “retinoic acid receptor-related orphan receptor γt” (RORγt) were found to play an important role in regulating inflammatory responses in the intestine (Ohnmacht et al., 2015; Sefik et al., 2015; Tanoue et al., 2016). On the other hand, gut bacteria capable of causing inflammatory effects can also be present. This is exemplified by segmented filamentous bacteria (SFB), commensal murine gut microbes that have been strongly correlated with an upregulated Th17 response in the small intestine (Wu et al., 2010) (Fig. 5). In humans, an analogous process appears to be mediated by Bifidobacterium adolescentis, as this bacterium was shown to be capable of inducing, alone, Th17 cell production in the small intestine of mice (Tan et al., 2016) (Fig. 5B). Another important example comes from the bacterial genus Prevotella. A relative increase of Prevotella species in the gut microbiota has been correlated with the reduction of the Bacteroides populations (Scher et al., 2013). Accordingly, Prevotella species potentially suppress the anti-inflammatory effect of
Bacteroides species such as B. fragilis. Prevotella copri, in fact, has been linked to an
inflammatory response via Th17 cells in the context of RA (Maeda et al., 2016). Firstly, its prevalence in the gut microbiome of new-onset untreated RA patients was reported to be significantly more abundant, when compared to healthy individuals (Scher et al., 2013). Secondly, P. copri has been proven to produce Pc-p27, a protein that induces reactivity of T helper 1 (Th1) cells, another RA-correlated proinflammatory T cell subset (Chen et al.,
2017), in 42% of new-onset RA patients (Pianta et al., 2017) (Fig. 5B). Lastly, antibodies against P. copri were found to be extremely specific for rheumatoid arthritis, suggesting a role for this gut bacterium in the pathogenesis of RA (Pianta et al., 2017). On the other hand, another member of the same genus, Prevotella histicola, has been found to suppress the inflammation in a collagen-induced murine arthritis model (Marietta et al., 2016), suggesting that different Prevotella species may have different effects on the pathogenesis of RA.
Remarkably, gut bacteria are also capable of forming biofilms, another phenomenon capable of altering the Th17/Tregs equilibrium. Specific biofilm components, such as amyloid fibrils (Taglialegna et al., 2016) and DNA, which serve as building blocks in the biofilm formation, have been implicated in autoimmunity via TLR recognition and subsequent Th17 activation (Dalpke et al., 2006; Schlafer et al., 2017). Indeed, the study of Gallo et al. demonstrated that a specific type of amyloid fibrils, the curli, when irreversibly associated with bacterial DNA, triggered the production of autoantibodies in a murine lupus model, suggesting a role for chronic biofilm-producing enteric infections in the pathogenesis of SLE and other autoimmune diseases (Gallo et al., 2015). Beside directly inducing autoantibody production, curli produced by enteric bacteria also activate the so-called NLRP3 inflammasome in murine macrophages, leading to production of inflammatory interleukin IL-1β (Rapsinski et al., 2015). This cytokine has been implicated in the differentiation of Th17 cells (Wang et al., 2012; Noack & Miossec, 2014) (Fig. 5B). Additionally, another study showed that upon entry of the enteric pathogen Salmonella enterica in the intestines of a colitis mouse model, the produced curli activated TLR2 responses, contributing to Th1 and Th17 activation, thereby promoting gut inflammation (Nishimori et al., 2012). Nevertheless, a recent study using a mouse model of colitis also revealed that epithelial barrier integrity, which is essential for gut homeostasis, is promoted by recognition of enteric bacterial curli by TLR2 (Oppong et al., 2015).Therefore, whether bacterial curli exert a protective or pathogenic role must depend on other factors, such as the phagocytic capacity of macrophages to digest and clear these biofilm components (Oppong et al., 2015). In this respect, our understanding of bacterial
amyloids and their roles in the gut microbiome and the pathogenesis of autoimmune diseases needs to be expanded. Remarkably, however, bacteria are not exclusive when it comes to amyloid production. Human amyloids have been extensively studied in relation to several neurodegenerative diseases, such as Alzheimer‟s and Parkinson‟s disease (Westermark et al., 2015). Similarly to the study of Gallo et al., other studies showed that nucleic acid-containing amyloid fibrils from human origin were also implicated in SLE pathogenesis (Di Domizio et al., 2012). In this context, it is particularly noteworthy that the microbiome was recently purported to play a potential role in the production of amyloids by human epithelial cells, specifically the acute phase protein serum amyloid A (SAA), which was shown to modulate neutrophil migratory behaviors and to induce Th17 responses in the gut (Kanther et al., 2014; Sano et al., 2015). Altogether, pathogenic and commensal members of the gut microbiome have been shown to produce amyloid fibrils in biofilms, while at the same time they are capable of stimulating SAA production by human intestinal cells, promoting a proinflammatory state (Fig. 5B).
Microbiome-based therapeutic management of RA
Despite the exponentially growing number of discoveries in the field of microbiomes, limited applications for new therapeutic avenues in the treatment of RA have been reported. However, pharmaceutical companies are expressing an increased interest in how to manipulate the microbiome to achieve positive health changes. With the understanding of the exact mechanisms by which specific microbes interact with one another and with the human host, future therapeutic strategies could aim at delivering specific bacteria to restore eubiosis, or ameliorate the effects of a dysbiotic microbiome. The microorganisms capable of conferring beneficial aspects to the host, when administered in adequate amounts, are termed probiotics (Bedaiwi & Inman, 2014; de Oliveira et al., 2017). An interesting example of this is the aforementioned case of P. histicola. This probiotic might prove itself a potential therapy for RA, as it was shown to suppress arthritis in humanized mice via mucosal regulation, more specifically the generation of Treg cells (Marietta et al., 2016). Recently, the
probiotics Lactobacillus paracasei and Lactobacillus casei have garnered special attention since they were shown to decrease inflammatory events by selectively degrading proinflammatory cytokines through their protease, lactoceptin (von Schillde et al., 2012). In the case of RA, it was reported that oral administration of L. casei resulted in decreased Th1 effector functions in a murine collagen-induced arthritis model (So et al., 2008). Similar observations were made in humans, where L. casei was given to RA patients who subsequently showed a significant decrease in proinflammatory cytokines, resulting in a lower disease score compared to untreated patients (Vaghef-Mehrabany et al., 2014). Another meaningful contribution to this field is the study of Vong et al., in which it was demonstrated that the probiotic bacterium Lactobacillus rhamnosus inhibited formation of pathogen-induced NETs (Vong et al., 2014). Vong et al. also demonstrated the differential capacity of different gut microbiome subsets to elicit neutrophil activation and NETosis (Vong
et al., 2015). Even though beneficial characteristics have been associated with the use of
probiotic bacteria, a full-fledged status as adjunctive therapy remains to be fully substantiated (Pineda Mde et al., 2011; Mohammed et al., 2017). Only few studies, in fact, investigated the possibility of translating the beneficial effects that probiotics have on RA animal models to humans (Schorpion & Kolasinski, 2017; Horta-Baas et al., 2017).
Considering the importance of a healthy gut microbiome in relation to autoimmune diseases, fecal microbial transplantation (FMT) has also been considered as a potential therapy for RA. However, while FMT proved highly effective for certain diseases, such as infectious colitis (van Nood et al., 2013), it still presents several challenges, especially in not yet established therapeutic applications. The positive results obtained by studies investigating the effects of FMT in autoimmune diseases such as irritable bowel disease, however, suggest a potential future for this technique in the therapeutic landscape of RA (Colman & Rubin, 2014; Scher et al., 2016a; El-Salhy & Mazzawi, 2018). Taken together, these results indicate that mechanistic studies on diverse commensal bacteria will very likely lead us to the discovery of new species involved in tissue homeostasis and these, in turn, might
possibly be used as a prevention treatment for autoimmune diseases such as RA. Nonetheless, since the gut microbiota is greatly influenced by multiple factors, the identification of microbiome-based approaches that can be used universally to treat or prevent a disease still remains a challenge. In particular, genomic makeup and diet can profoundly influence the microbiome composition, making them potential obstacles when devising a universal probiotic therapy. This sparks the need for more personalized approaches (Goodrich et al., 2014; Valdes et al., 2018). On the other hand, the possibility that a given microorganism might not be capable of surviving its passage throughout the stomach acids, or of effectively colonizing a target niche that is already occupied by an endogenous microbial population, poses a potentially greater challenge (Mimee et al., 2016; Dodoo et al., 2017). To overcome these hurdles, particular chassis bacteria may have to be chosen after extensive research on the microbial ecology of the target niche, or be re-engineered to better suit the therapeutic needs (Mimee et al., 2016). For example, recent studies in mice have shown that bacterial strains engineered to metabolize a peculiar dietary component are capable of engrafting themselves into the already established gut microbial community when administered in concomitance with the respective dietary component (Kearney et al., 2018; Shepherd et al., 2018). Another factor that should be accounted for when devising a microbiome-based therapy is the variation in host responses elicited by different strains of the same species. This is underscored by studies of Geva-Zatorsky et al., who demonstrated that certain Bacteroides species display strain-specific differential immunomodulatory capacities (Geva-Zatorsky et al., 2017). Considering these challenges, the use of small molecules mimicking the interaction between beneficial bacteria and the host could represent a better alternative for therapy (Garber, 2015). In this respect, an identified bacterial molecule of interest is the aforementioned PSA from B. fragilis, which was shown to induce the maturation of the host immune system and to elicit protective effects against colitis by promoting the production of Foxp3+ Treg cells (Mazmanian et al., 2005;
Round & Mazmanian, 2010). Based on these findings, a PSA-based oral therapy was created to treat autoimmune, inflammatory, and allergic diseases (Garber, 2015).
An alternative potential microbiome-based therapy for RA targets the pathogens implicated in the etiology of this disease, such as P. gingivalis. For example, a recent study in mice with experimental arthritis showed amelioration of the symptoms of both periodontitis and collagen-induced arthritis upon treatment with a FimA antibody (Jeong et al., 2018). This antibody, targeting the major fimbrillin protein of P. gingivalis, appeared to attenuate bacterial attachment and aggregation on the tested murine fibroblasts (Jeong et al., 2018). An additional potential avenue for therapy concerns the formation of biofilm by oral bacteria. New compounds capable of inhibiting quorum sensing molecules, such as AI-2, have been recently tested both in vivo and in vitro. Perhaps due to the novelty of this field, few studies have investigated the beneficial effects of quorum sensing inhibitors (Cho et al., 2016; Park
et al., 2017). However, the potential therapeutic effect they demonstrated in periodontitis,
which is mainly due to their capability to limit biofilm formation, shows some promise for eventual future applications in the treatment of RA. Nevertheless, with the important and very rapid advances in the “omics” field and the development of massive computational approaches, the translational potential of emerging therapeutic agents has become increasingly more evident. To support this, in the case of RA, Tieri et al. developed a multi-omic map to estimate the outcomes of novel therapies focusing on several aspects of RA, including immune responses mediated by the gut microbiome, leading to potential targets for RA treatment (Tieri et al., 2014). Overall, as the scientific community continues to elucidate the complex relationships between the microbiome and human health, numerous therapeutic targets are being identified which, in turn, may be efficiently tested in silico to predict, to some extents, the outcome of future clinical trials. Hopefully, advances as summarized above will bring us one step closer to the discovery of a „universal‟ therapy to prevent or treat RA and other autoimmune diseases.
Concluding Remarks
Over the past decade, the field of microbiome research was exponentially expanded by more and more studies demonstrating the profound impact of the microbiome on human
health and disease. Moreover, important technological approaches have advanced our understanding of how the microbiome and the host interact to maintain tissue homeostasis. Yet, major knowledge gaps still exist. For instance, the functional consequences of microbial biofilm formation need to be evaluated in more detail, because recent studies have provided evidence that biofilm-related phenomena and components influence the immune system. A second important area that needs to be explored more in depth concerns the microbial modulation of neutrophil activity, especially since neutrophils could represent a key source of autoantigens in RA patients. In particular, recent observations imply that the oral microbiome influences NETosis. However, further insight is needed to pinpoint the microbial populations responsible for altering neutrophil activities, potentially leading to the development of microbial therapies that restore neutrophil homeostasis. Luckily, novel „omics‟-based technologies to study the microbiome and its interactions with the host have recently been developed. These can now be exploited to overcome the limitations of classical biochemical and immunological approaches. Moreover, innovative computational tools will allow us to predict and verify the outcomes of potential therapies. As a consequence, we will surely experience, in the near future, a boost of preventive and therapeutic approaches to promote, balance, or restore the beneficial interactions between the microbiome and the human body.
Ethics approval
Figure 4 was recorded in the context of a previous study that received Institutional Review Board approval from the Medical Ethics Committee of the University Medical Center Groningen (METc UMCG 2007.195). This study was performed in accordance with the guidelines of the Declaration of Helsinki and the institutional regulations, and all samples were anonymized.
Availability of data and materials
Not applicable
Competing interests
The authors declare that they have no financial and non-financial competing interests in relation to the documented research.
Funding
This work was supported by the Graduate School of Medical Sciences of the University of Groningen [to M.d.T. and G.G] and the Center for Dentistry and Oral Hygiene of the University Medical Center Groningen [to G.G., A.J.v.W.].
Authors’ contribution
M. du Teil Espina and G. Gabarrini drafted the manuscript. H.J.M. Harmsen provided Figure 3 and J. Westra provided Figure 4. A.J. van Winkelhoff and J.M. van Dijl supervised the project. All authors, critically revised the manuscript, gave final approval and agree to be accountable for all aspects of the review.
Acknowledgements
We thank Vincent Zijnge for providing the image in Figure 3, and Menke de Smit, Tim Stobernack, Elisabeth Brouwer, Arjan Vissink, Peter Heeringa, and Sasha Zhernakova for stimulating discussions.
References
Aagaard K, Ma J, Antony KM, Ganu R, Petrosino J & Versalovic J (2014) The placenta harbors a unique microbiome Sci Transl Med 6: 237ra65.
Abramson SB, Given WP, Edelson HS & Weissmann G (1983) Neutrophil aggregation induced by sera from patients with active systemic lupus erythematosus Arthritis Rheum 26: 630-636.
Al-Katma MK, Bissada NF, Bordeaux JM, Sue J & Askari AD (2007) Control of periodontal infection reduces the severity of active rheumatoid arthritis J Clin Rheumatol 13: 134-137.
Arimatsu K, Yamada H, Miyazawa H et al. (2014) Oral pathobiont induces systemic inflammation and metabolic changes associated with alteration of gut microbiota Sci Rep 4: 4828.
Arkema EV, Goldstein BL, Robinson W, Sokolove J, Wagner CA, Malspeis S, Rosner B, Grodstein F, Karlson EW & Costenbader KH (2013) Anti-citrullinated peptide autoantibodies, human leukocyte antigen shared epitope and risk of future rheumatoid arthritis: a nested case-control study Arthritis Res Ther 15: R159.
Atarashi K, Suda W, Luo C et al. (2017) Ectopic colonization of oral bacteria in the intestine drives TH1 cell induction and inflammation Science 358: 359-365.
Avouac J, Gossec L & Dougados M (2006) Diagnostic and predictive value of anti-cyclic citrullinated protein antibodies in rheumatoid arthritis: a systematic literature review Ann Rheum Dis 65: 845-851.
Bedaiwi MK & Inman RD (2014) Microbiome and probiotics: link to arthritis Curr Opin Rheumatol 26: 410-415.
Belibasakis GN, Thurnheer T & Bostanci N (2013) Interleukin-8 responses of multi-layer gingival epithelia to subgingival biofilms: role of the "red complex" species PLoS One 8: e81581.
Belkaid Y & Hand TW (2014) Role of the microbiota in immunity and inflammation Cell 157: 121-141.
Blanc V, O'Valle F, Pozo E, Puertas A, Leon R & Mesa F (2015) Oral bacteria in placental tissues: increased molecular detection in pregnant periodontitis patients Oral Dis 21: 905-912.
Brusca SB, Abramson SB & Scher JU (2014) Microbiome and mucosal inflammation as extra-articular triggers for rheumatoid arthritis and autoimmunity Curr Opin Rheumatol 26: 101-107.
Buford TW (2017) (Dis)Trust your gut: the gut microbiome in age-related inflammation, health, and disease Microbiome 5: 80.
Carrion J, Scisci E, Miles B, Sabino GJ, Zeituni AE, Gu Y, Bear A, Genco CA, Brown DL & Cutler CW (2012) Microbial carriage state of peripheral blood dendritic cells (DCs) in chronic periodontitis influences DC differentiation, atherogenic potential J Immunol 189: 3178-3187. Chen B, Sun L & Zhang X (2017) Integration of microbiome and epigenome to decipher the pathogenesis of autoimmune diseases J Autoimmun 83: 31-42.
Cho I & Blaser MJ (2012) The human microbiome: at the interface of health and disease Nat Rev Genet 13: 260-270.
Cho YJ, Song HY, Ben Amara H et al. (2016) In Vivo Inhibition of Porphyromonas gingivalis
Growth and Prevention of Periodontitis With Quorum-Sensing Inhibitors J Periodontol 87: 1075-1082.
Chukkapalli S, Rivera-Kweh M, Gehlot P, Velsko I, Bhattacharyya I, Calise SJ, Satoh M, Chan EK, Holoshitz J & Kesavalu L (2016) Periodontal bacterial colonization in synovial tissues exacerbates collagen-induced arthritis in B10.RIII mice Arthritis Res Ther 18: 161. Chung WO, Park Y, Lamont RJ, McNab R, Barbieri B & Demuth DR (2001) Signaling system
in Porphyromonas gingivalis based on a LuxS protein J Bacteriol 183: 3903-3909.
Colman RJ & Rubin DT (2014) Fecal microbiota transplantation as therapy for inflammatory bowel disease: a systematic review and meta-analysis J Crohns Colitis 8: 1569-1581.
Correa JD, Calderaro DC, Ferreira GA, Mendonca SM, Fernandes GR, Xiao E, Teixeira AL, Leys EJ, Graves DT & Silva TA (2017) Subgingival microbiota dysbiosis in systemic lupus erythematosus: association with periodontal status Microbiome 5: 34.
Corsiero E, Bombardieri M, Carlotti E, Pratesi F, Robinson W, Migliorini P & Pitzalis C (2016) Single cell cloning and recombinant monoclonal antibodies generation from RA synovial B cells reveal frequent targeting of citrullinated histones of NETs Ann Rheum Dis 75: 1866-1875.
Dalpke A, Frank J, Peter M & Heeg K (2006) Activation of toll-like receptor 9 by DNA from different bacterial species Infect Immun 74: 940-946.
Darrah E & Andrade F (2017) Rheumatoid arthritis and citrullination Curr Opin Rheumatol.
